| Summary: | 碩士 === 長庚大學 === 資訊管理學系 === 99 === Early diagnosis and treatment of breast cancer can effectively decrease the mortality rate. Recently, ultrasound examination plays an important role in the field of breast cancer diagnosis because of its non-invasive, low price, and convenience.
To promote the classification accuracy and decrease training time of an ultrasound breast tumor image computer-aided diagnosis system, we proposed an approach which combined feature selection and parameter setting simultaneously in this paper.
Before the classification, all the breast tumors were segmented automatically by a level set method. Then, the texture features and shape features were first extracted following the use of an artificial immune system algorithm to detect significant features and determine the near-optimal parameters for the support vector machine to identify the tumor as benign or malignant.
The experiment shows that the accuracy of the proposed system for classifying breast tumors is 95.24%, the sensitivity is 97.78%, the specificity is 93.33%, the positive predictive value is 91.67%, and the negative predictive value is 98.25%. It is proved that the use of an artificial immune system algorithm can promote the classification accuracy and decrease training time of the support vector machine.
|
|---|
